Cobol Scanner should recover when *, identification Area is present in column 72 of COBOL code

At my organization, some COBOL programs in production import copybooks, and those copybooks may themselves include other copybooks. For example, the program flow might look like:

olila12p.cob → DPBAUTW0.cpy → DPSINSCC.cpy

However, the Manta COBOL scanner encounters issues when processing COPY statements that contain an asterisk (*) in column 72. This character is typically used to denote comments, but in these cases, it appears within the code section rather than the comment section. This causes the scanner to fail.

Additionally, we found that some Cobol file has an identification Area (00022400) which starts from column 72(Ref: TS020140334) and this is correct Application code. The COBOL scanner should not fail in such cases.

Example from olila12p.cob:

006400*01 - WS-DPSINSCC. *
006500 COPY DPSINSCC. *
006600*----------------------------------------------------------------*
006700*DATA DEFINITION STARTS HERE *
006800*----------------------------------------------------------------*
006900 02 DPBAUTW0M-RECORD.
007000 COPY MISPREA REPLACING
007100 ==:MICNT:== BY ==DPAUTW0M==.
007200*
007300 02 DPBAUTW0-RECORD.

Expected Behavior:

Manta should detect and log such unexpected scenarios where COPY statements are affected by misplaced asterisks.

These issues should not prevent the generation of lineage information in the Manta Flow Viewer. The full lineage of the file should still be displayed, even if such anomalies are present.

Workaround:

In cases when identification Area present in column 72 of cobol code, current workaround is to use replace.csv.

In cases when * is present in column 72 of Cobol code, attempt made with replace.csv results in other errors in the code.

Impact:

We found this issue in two of the Cobol programs, given that we only scanned two Cobol application so far.